Abstract
This thesis considers the energy-efficient precoding matrix design for a relay-aided multiuser downlink multiple-input single-output (MISO) wireless system. This problem is nonconvex and complicated in general and has drawn extensive attention, but few effective and efficient algorithms have been reported. In this thesis, the precoders of the base station (BS) and the relay station (RS) both equipped with multiple antennas are designed to maximize the transmit energy efficiency (EE), defined as the ratio between the system sum rate and the total power consumption, under respective quality-of-service (QoS) constraints of singleantenna users and the transmit power constraints on the BS and the RS. In view of the fact that the associated optimization problem is a nonconvex fractional programming, a successive Dinkelbach and convex approximation (SDCA) algorithm with convergence guaranteed is proposed to cope with the problem. Some simulation results are provided to demonstrate the effectiveness of the proposed SDCA algorithm, and significant EE improvement as the number of antennas at the BS and the RS increases (i.e., more spatial degrees of freedom). Finally, some conclusions are provided.